Roar Digital Search Radar: False Positives on Our Site

What happened

On 13 October 2026 we ran the free Search Radar visibility report from Roar Digital on aiwebpageseo.com. It promises "your rankings, your competitors, your AI visibility. All real data. All in one place." It gave us 42 out of 100 and the label "Weak visibility".

That score rests on one decision the tool made at the start: that we are a local SEO and website audit service in Sheffield. We are not in Sheffield, and we are not a local service. Almost everything after that follows from the mistake. This post goes through the report section by section and checks each claim against our live site.

The short version

• Wrong town. It "detected" us in Sheffield. Our business address is in Chesterfield, and our home page names Chesterfield seven times and Sheffield not once.
• Wrong market. All three "searches that matter" were local Sheffield searches. Our audits run online for any website, anywhere, and our structured data says we serve the whole United Kingdom.
• Wrong rivals. Its "biggest rival" is a cyber security firm. The others ahead of us include a students' union.
• "Analytics installed ✗" Google Analytics is installed. It only loads after a visitor accepts analytics cookies, so a scanner that never clicks "accept" cannot see it.
• "No information on customer reviews." Our Trustpilot profile is linked from the home page and listed in our structured data.
• "Little presence beyond the website." Our structured data lists 18 official profiles and directory listings.
• Where it leads. The AI section ends with a link to Roar Digital's own service page, and the "full report" is emailed to you.

Finding 1: "Detected: SEO and website audit services in Sheffield"

What the report said. Under the score, the report states what it detected: "SEO and website audit services in Sheffield".

What the site says. The Organization and LocalBusiness markup on our home page gives our address in Chesterfield. The word Sheffield does not appear anywhere on the page. The same markup gives our service area as the United Kingdom, because the audits run online and work on any public website.

Why it matters. Chesterfield is about 12 miles from Sheffield, so a tool that guesses location from the nearest big city might land there. But a visibility report is only as good as its idea of who you are. Get the town and the type of business wrong, and every ranking, rival and recommendation that follows is measured against the wrong target.

Finding 2: "Searches that matter for your business"

What the report said. It tested four searches and reported:

• "website security audit Sheffield": not on page 1, behind aptive.co.uk and basaltcyber.com
• "SEO accessibility audit services Sheffield": not on page 1, behind su.sheffield.ac.uk and webtonic.io
• "schema markup audit Sheffield": not on page 1, behind get-found.co.uk and seoclarity.net
• "aiwebpageseo": number 1

What is wrong with that. Three of the four searches add a city name to a service that has nothing to do with location. Someone who wants to check their own website's security headers or schema does not search for a provider in a particular city: they search for a tool, and they use it from wherever they are. A city search is the right test for a plumber, a dentist or a high-street agency that visits clients. It is the wrong test for an online audit platform.

One of the sites "ahead of" us for the accessibility search is su.sheffield.ac.uk, the University of Sheffield Students' Union. That tells you the search is matching the word Sheffield, not a competing service.

Finding 3: "Biggest rival: aptive.co.uk"

What the report said. aptive.co.uk "appears across the searches that matter to you".

What that is. Aptive is a cyber security consultancy whose services include application security assessments, vulnerability assessments and penetration testing. It is a rival only because the report chose local security-audit searches. A penetration test, where specialists actively try to break into your systems, is a different service from our passive Security Audit, which reads only what any visitor's browser can already see. The "biggest rival" is a product of the wrong searches, not of the market we work in.

Finding 4: "Analytics installed ✗"

What the report said. In the Website health checklist, every item passed except one: "Analytics installed", marked with a cross.

What the site does. Google Analytics is installed. It is loaded only after a visitor accepts analytics cookies in our cookie banner, and the code on the page says so. Until then, nothing is sent to Google. Under the Privacy and Electronic Communications Regulations (PECR), storing or reading information on a visitor's device needs consent unless an exception applies. Since the Data (Use and Access) Act 2025, the Information Commissioner's Office guidance includes an exception for some analytics, as long as visitors are told and can easily opt out. Asking first is still the most cautious way to run analytics, and it is a choice we made on purpose.

Why the scan got it wrong. A scanner that loads the page without clicking "accept" will never see analytics that wait for consent. That is how consent-first analytics is meant to work, not a missing feature. A health check that marks a site down for asking visitors first is telling owners to stop asking. With that item corrected, our Website health result is 12 out of 12.

Finding 5: "No information on customer reviews"

What the report said. Listed under "Where AI sees gaps".

What the site has. Our Trustpilot profile is linked from the home page and is one of the profiles listed in the sameAs property of our Organization markup, which is the standard way to tell search engines and AI systems which profiles belong to a business.

Finding 6: "Little presence beyond the website"

What the report said. Also listed under "Where AI sees gaps".

What the site has. Our Organization and Person markup each list 18 official profiles: LinkedIn, X, Facebook, Instagram, YouTube, TikTok, Threads, Reddit, Hacker News, Slack, our Google Business Profile and Maps listing, Trustpilot and several UK business directories. The same report ticked "Social profiles" as a pass in its own Website health checklist, a few lines above.

Finding 7: what the AI section says, and where it leads

The "What AI says about your business" section repeats the Sheffield mistake: it says AI sees us as focused on audit services "particularly in Sheffield". It then answers "Would AI recommend you?" with a caution about "limited visibility and competition in the local market", a market we do not serve.

Directly under that answer is a link titled "How to make AI more likely to recommend you". It goes to a page on roardigital.co.uk about Roar Digital's own services. The report ends with "Report on its way: We'll put your full report together and email it over shortly." A free report that finds weak visibility and then points to the publisher's services is a sales funnel. That is a fair business model, but it is worth knowing when you read the score.

Finding 8: the score

The report does not say how 42 out of 100 is calculated. Its own checklist passed 11 of 12 website health items (12 of 12 once analytics is counted correctly), it found the site ranks first for its own name, and it said AI recognises the business. The low score comes from the three Sheffield searches, which are the wrong searches. A score built on searches a business does not compete for cannot tell that business how visible it is.

The questions it suggested, answered

The report did suggest four questions people ask. They are good questions, so here are short answers. This is general information, not legal advice.

How much does an accessibility audit cost in the UK?

It depends on the kind of audit. An automated audit checks a page against the criteria a tool can test. Our Accessibility Audit costs £2.49, checks the page against WCAG 2.2 Level A and AA as visitors see it, checks two more pages of the site, and lists which of the 55 criteria need a person to check. A manual audit by an accessibility specialist covers those criteria too, using screen readers and keyboard testing, and is priced by the specialist for the size of the site.

How much does a WCAG audit cost?

The same answer applies: a WCAG audit is an accessibility audit measured against the Web Content Accessibility Guidelines. Ask any provider which version and level they test against (WCAG 2.2 Level AA is the current benchmark), how many pages are covered, and whether the report lists the criteria that were not tested.

Is WCAG a legal requirement in the UK?

For public sector bodies, yes in effect: the Public Sector Bodies Accessibility Regulations 2018 apply, and GOV.UK guidance says their websites must meet WCAG 2.2 AA and publish an accessibility statement. For businesses, the Equality Act 2010 requires service providers, including through a website, to make reasonable adjustments so disabled people are not put at a substantial disadvantage. The Act does not name a technical standard, but WCAG is the standard used to judge whether a website is accessible.

Can you be sued in the UK for your website not being accessible?

The Equality Act 2010 applies to services provided through websites, and a disabled person who is put at a substantial disadvantage by a failure to make reasonable adjustments can bring a claim under the Act. Meeting WCAG 2.2 Level AA is the clearest way to show a website has been made accessible.

How to read any visibility report

• Check what it thinks you are. If the business type or location is wrong, stop there: everything after it is measured against the wrong target.
• Check the searches. Are they the words your customers actually use? An online service tested on city searches will always look invisible.
• Check the rivals. If a "competitor" does not sell what you sell, the searches are wrong.
• Check the crosses. A "missing" tracker may simply be waiting for cookie consent.
• Check where the links go. If the fix is the publisher's own service, read the score with that in mind.

Sources

• ICO: Guidance on the use of storage and access technologies
• GOV.UK: Understanding accessibility requirements for public sector bodies
• Equality Act 2010
• schema.org: sameAs
• schema.org: areaServed

Frequently asked questions

Is Roar Digital's Search Radar report accurate?

On our site it was not. It decided we were a local audit service in Sheffield, when we are an online platform based in Chesterfield that serves the whole UK. It then tested three Sheffield searches, named a cyber security firm as our biggest rival, marked analytics as missing when it loads after cookie consent, and said we had no reviews and little presence beyond the website, when our Trustpilot and 18 official profiles are listed on the home page.

Why would a visibility report say analytics is not installed?

Because many sites only load analytics after a visitor accepts cookies. A scanner that loads the page without clicking accept never sees the analytics script, so it reports it as missing. That is how consent-first analytics is meant to work. Check your cookie banner and your analytics account before acting on a report that says analytics is missing.

Should an online service be judged on local city searches?

No. City searches are the right test for businesses that serve customers in one area, such as plumbers, dentists or local agencies. An online tool that anyone can use from anywhere competes on searches for the tool itself, not on searches that include a town name, so a report built on city searches will always make it look invisible.

Is WCAG a legal requirement in the UK?

For public sector bodies, GOV.UK guidance says websites must meet WCAG 2.2 AA and publish an accessibility statement under the Public Sector Bodies Accessibility Regulations 2018. For businesses, the Equality Act 2010 requires reasonable adjustments so disabled people are not put at a substantial disadvantage, including through a website. The Act names no technical standard, but WCAG is the standard used to judge whether a website is accessible.